Merge branch 'pr-democracynow' of https://github.com/atomicdryad/youtube-dl into...

[youtube-dl] / youtube_dl / extractor / fktv.py

diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py
index c2aa23aa28f7c257d697ee2567f3b64848dfbcb8..40ea2789525ab720beb9e6e42a096682a9581027 100644 (file)
--- a/youtube_dl/extractor/fktv.py
+++ b/youtube_dl/extractor/fktv.py
@@ -6,6 +6,7 @@ from .common import InfoExtractor
 from ..utils import (
     clean_html,
     determine_ext,
+    ExtractorError,
 )
 
 
@@ -20,23 +21,35 @@ class FKTVIE(InfoExtractor):
             'id': '1',
             'ext': 'mp4',
             'title': 'Folge 1 vom 10. April 2007',
+            'thumbnail': 're:^https?://.*\.jpg$',
         },
     }
 
     def _real_extract(self, url):
         episode = self._match_id(url)
 
-        webpage = self._download_webpage('http://fernsehkritik.tv/folge-%s/play' % episode, episode)
-        title = clean_html(self._html_search_regex('<h3>([^<]+?)</h3>', webpage, 'title'))
-        matchs = re.search(r'(?s)<video[^>]*poster="([^"]+)"[^>]*>(.*?)</video>', webpage)
-        if matchs:
-            poster, sources = matchs.groups()
-            urls = re.findall(r'(?s)<source[^>]*src="([^"]+)"[^>]*>', sources)
-            if sources:
-                formats = [{'url': url, 'format_id': determine_ext(url)} for url in urls]
-                return {
-                    'id': episode,
-                    'title': title,
-                    'formats': formats,
-                    'thumbnail': poster,
-                }
+        webpage = self._download_webpage(
+            'http://fernsehkritik.tv/folge-%s/play' % episode, episode)
+        title = clean_html(self._html_search_regex(
+            '<h3>([^<]+)</h3>', webpage, 'title'))
+        matches = re.search(
+            r'(?s)<video(?:(?!poster)[^>])+(?:poster="([^"]+)")?[^>]*>(.*)</video>',
+            webpage)
+        if matches is None:
+            raise ExtractorError('Unable to extract the video')
+
+        poster, sources = matches.groups()
+        if poster is None:
+            self.report_warning('unable to extract thumbnail')
+
+        urls = re.findall(r'<source[^>]+src="([^"]+)"', sources)
+        formats = [{
+            'url': furl,
+            'format_id': determine_ext(furl),
+        } for furl in urls]
+        return {
+            'id': episode,
+            'title': title,
+            'formats': formats,
+            'thumbnail': poster,
+        }